System And/Or Method for Image Tamper Detection

1. A method comprising: representing an image as image data; and detecting, by a processing system, a tampering in said image based, at least in part, on classification features generated based, at least in part, on said image data according to a Hilbert-Huang transformation.

2. The method of claim 1 , wherein said detecting said tampering further comprises: extracting a structure component from said image data; deriving one or more intrinsic mode functions from said structure component; and detecting said tampering in said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component.

3. The method of claim 2 , wherein said detecting said tampering further comprises: predicting features of said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component; and comparing said predicted features with actual features of said image to determine an error.

4. The method of claim 1 , and further comprising generating said classification features based, at least in part, on one or more intrinsic mode functions derived from a fine texture component extracted from said image data.

5. The method of claim 1 , and further comprising generating said classification features based, at least in part, on one or more intrinsic mode functions derived from image data representing an authentic image.

6. The method of claim 1 , wherein said image data comprises an array including a plurality of rows and a plurality of columns, the method further comprising generating said classification features based, at least in part, on 1-D Hilbert-Huang transforms of individual rows to provide associated IMFs and Hilbert spectrums associated with said individual rows.

7. The method of claim 6 , the method further comprising generating said classification features based, at least in part, on 1-D Hilbert-Huang transforms of individual columns to provide associated IMFs and Hilbert spectrums associated with said individual columns.

8. The method of claim 7 , wherein said generating said classification features further comprises generating 16-D features based, at least in part, on said IMFs and Hilbert spectrums associated with said individual rows and columns.

9. The method of claim 1 , and further comprising detecting said tampering in said image based, at least in part, on classification features generated based, at least in part, on statistical moments of characteristic functions of a prediction-error image, test image, and associated wavelet subbands.

10. The method of claim 1 , wherein said tampering comprises one or more splices in said image.

11. The method of claim 1 , further comprising: deriving one or more sets of classification features based on a set of test image data, the classification features generated at least in part based on a Hilbert-Huang transformation; and training a classifier to perform said detecting based on the one or more sets of classification features.

12. An apparatus comprising: a computing platform, said computing platform being configured to: represent an image as image data; and detect a tampering in said image based, at least in part, on classification features generated based, at least in part, on said image data according to a Hilbert-Huang transformation.

13. The apparatus of claim 12 , wherein said computing platform is further adapted to: extract a structure component from said image data; derive one or more intrinsic mode functions from said structure component; and detect said tampering in said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component.

14. The apparatus of claim 13 , wherein said computing platform is further adapted to: predict features of said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component; and compare said predicted features with actual features of said image to determine an error.

15. The apparatus of claim 12 , wherein said computing platform is further adapted to generate said classification features based, at least in part, on one or more intrinsic mode functions derived from a fine texture component extracted from said image data.

16. The apparatus of claim 12 , wherein said computing platform is further adapted to generate said classification features based, at least in part, on one or more intrinsic mode functions derived from image data representing an original image.

17. The apparatus of claim 12 , wherein said image data comprises an array including a plurality of rows and a plurality of columns, and wherein said computing platform is further adapted to generate said classification features based, at least in part, on one or more 1-D Hilbert-Huang transforms of individual rows to provide associated IMFs and Hilbert spectrums associated with said individual rows.

18. The apparatus of claim 17 , wherein said computing platform is further adapted to generate said classification features based, at least in part, on one or more 1-D Hilbert-Huang transforms of individual columns to provide associated IMFs and Hilbert spectrums associated with said individual columns.

19. The apparatus of claim 18 , wherein said computing platform is further adapted to generate 16-D features based, at least in part, on said IMFs and Hilbert spectrums associated with said individual rows and columns.

20. The apparatus of claim 12 , wherein said computing platform is further adapted to detect said tampering in said image based, at least in part, on classification features generated based, at least in part, on statistical moments of characteristic functions of a prediction-error image, test image, and associated wavelet subbands.

21. The apparatus of claim 12 , wherein said tampering comprises one or more splices in said image.

22. The apparatus of claim 12 , further comprising: an image classifier to detect said tampering, the image classifier configured to receive classification features generated from a set of test image data, the classification features derived at least in part based on a Hilbert-Huang transformation, and to transform its configuration based on the classifications.

23. The apparatus of claim 12 , wherein said computing platform comprises one or more processors.

24. The apparatus of claim 12 , further comprising a reception device configured to receive the image.

25. An article of manufacture including: a storage medium having machine-readable instructions stored thereon, the instructions comprising: instructions for representing an image as image data; and instructions for detecting tampering in said image based, at least in part, on classification features generated based, at least in part, on said image data according to a Hilbert-Huang transformation.

26. The article of claim 25 , wherein the instructions further comprise: instructions for extracting a structure component from said image data; instructions for deriving one or more intrinsic-mode functions from said structure component; and instructions for detecting said tampering in said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component.

27. The article of claim 26 , wherein the instructions further comprise: instructions for predicting features of said image based, at least in part, on said one or more intrinsic mode functions derived from said structure component; and instructions for comparing said predicted features with actual features of said image to determine an error.

28. The article of claim 25 , wherein the instructions further comprise: instructions for generating said classification features based, at least in part, on one or more intrinsic mode functions derived from a fine texture component extracted from said image data.

29. The article of claim 25 , wherein the instructions further comprise: instructions for generating said classification features based, at least in part, on one or more intrinsic mode functions derived from image data representing an original image.

30. The article of claim 25 , wherein said image data comprises an array including a plurality of rows and a plurality of columns, and wherein the instructions further comprise: instructions for generating said classification features based, at least in part, on one or more 1-D Hilbert-Huang transforms of individual rows to provide associated IMFs and Hilbert spectrums associated with said individual rows.

31. The article of claim 30 , wherein the instructions further comprise: instructions for generating said classification features based, at least in part, on one or more 1-D Hilbert-Huang transforms of individual columns to provide associated IMFs and Hilbert spectrums associated with said individual columns.

32. The article of claim 31 , wherein the instructions further comprise: instructions for generating 16-D features based, at least in part, on said IMFs and Hilbert spectrums associated with said individual rows and columns.

33. The article of claim 25 , wherein the instructions further comprise: instructions for detecting said tampering in said image based, at least in part, on classification features generated based, at least in part, on statistical moments of characteristic functions of a prediction-error image, test image, and associated wavelet subbands.

34. The article of claim 25 , wherein said tampering comprises one or more splices in said image.

35. An apparatus comprising: means for representing an image as image data; and means for detecting a tampering in said image based, at least in part, on classification features generated based, at least in part, on said image data according to a Hilbert-Huang transformation.